Pomegra
Meticulous raised $15M in Series A funding led by Chemistry, with Menlo Ventures joining, to test AI-written code autonomously before it ships to production.
Key Takeaways
- Meticulous closed a $15M Series A led by Chemistry, with Menlo Ventures participating. Valuation is undisclosed.
- Total funding reaches $19M, including a $4M seed round from January 2024.
- Customers include Notion, ElevenLabs, Dropbox, Wiz and LaunchDarkly. The company says ARR grew fivefold year over year.
Lead
Meticulous, a US startup that automatically generates and maintains end-to-end tests for web applications, has raised a $15M Series A in 2026. Chemistry led the round. Menlo Ventures participated, alongside angels including Lachy Groom, Jason Warner, Arash Ferdowsi, Scott Belsky, Guillermo Rauch, Calvin French-Owen, Caitlin Colgrove and Jason Ginsberg. The company did not disclose a valuation. It plans to spend the money on research and development, backend and performance testing, and a larger team.
What Does Meticulous Actually Do?
Meticulous records how people use an application during development, then turns those sessions into tests that run against each code change. It compares the app's behavior before and after an update, and its deterministic browsers can flag differences as small as a single pixel. Engineers, product managers and designers can all see what a change affects.
The pitch is that teams do not write or maintain the tests by hand. The platform scans the codebase, finds edge cases and keeps thousands of user flows current as the product changes. Today it covers the frontend only. Backend and full-stack coverage are planned.
Who Are the Founders and Customers?
The company was founded in 2021 by brothers Gabriel Spencer-Harper, a former Dropbox software engineer, and Quentin Spencer-Harper, who spent a decade leading frontend engineering on Palantir's Foundry product. Their customer list is short but recognizable: Notion, ElevenLabs, Dropbox, Wiz and LaunchDarkly.
A LaunchDarkly engineering director framed the problem this way: code generation is relatively inexpensive, but code review becomes the bottleneck. That is the gap Meticulous is aiming at. Coding assistants have raised the volume of pull requests, and human reviewers have not scaled with them.
Why Does This Round Matter for AI Coding Tools?
It matters because investors are placing money on verification rather than generation. Writing code has become cheap, and the open question is how teams trust what an agent produced. A testing layer that runs without manual upkeep is one answer, and a Series A with a seed-stage check history suggests investors see it as an early-stage bet, not a proven category.
The numbers are modest. The $15M follows a $4M seed from January 2024, and the company has about 20 employees. The fivefold ARR growth claim comes from the company, and it came without a base figure, so the revenue scale is unknown. Fivefold growth from a small base is common at this stage.
The angel list also tells a story. Founders and executives from Vercel, Cursor, Poolside, Dropbox and Hex are the people most exposed to AI-generated code in production, and they are backing a tool that polices it.
How Crowded Is the Field?
The category is busy and well funded. Harness, valued above $3.7B, sells software delivery tooling with testing features. CodeRabbit and Qodo focus on AI code review. Diffblue has raised more than $50M for automated unit test generation. SolveAI recently raised $50M.
Most of those rivals read or generate code and then reason about it. Meticulous works from recorded user behavior and replays it, which gives it a different failure profile. It will catch visual and behavioral regressions that a review model might miss, but it only sees flows that someone has exercised. Coverage of rarely used paths depends on how much real usage the system observes.
What Comes Next for Meticulous?
The company expects to grow from about 20 people to 30-40 within a year. It plans to hire forward-deployed engineers who sit with enterprise customers, a sign that large accounts are asking for hands-on integration work. It also plans to extend coverage into backend and performance testing, where deterministic replay is harder than in a browser.
Two tests lie ahead. The first is whether it can move beyond frontend checks without losing the low-maintenance promise. The second is whether bigger platforms, which already bundle testing with code hosting and delivery pipelines, build equivalent features before Meticulous reaches scale.
Outlook
Meticulous has $19M in total funding, five named customers of note and a thesis that review, not writing, is the constraint on AI-assisted development. The round is small by current AI standards, and the valuation stays private. Over the next year, backend expansion, hiring pace and whether the fivefold growth holds against well-capitalized competitors will show whether autonomous testing becomes a standalone product or a feature inside larger platforms.
